A predictive maintenance method and device for elevators based on the Internet of Things and machine learning

By building an elevator condition monitoring system using the Internet of Things and machine learning, and using the random forest algorithm to establish an elevator fault prediction model, the problems of large workload, uneven technical level and high cost of elevator maintenance have been solved. Predictive maintenance of elevator faults has been realized, improving elevator safety and maintenance efficiency.

CN118545590BActive Publication Date: 2026-11-17浪潮智慧城市科技有限公司
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202410762105.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-13
Publication Date
2026-11-17
Estimated Expiration
2044-06-13

AI Technical Summary

Technical Problem

The current elevator maintenance work is labor-intensive, technical levels are uneven, and costs are high. Traditional maintenance models are inefficient and difficult to achieve precision and comprehensiveness.

Method used

An elevator status monitoring system is built based on the Internet of Things. Data is collected through sensors, and a machine learning algorithm is used to build an elevator fault prediction model. The random forest algorithm is used to predict the health status and remaining life of elevator internal components, and historical data is combined for model training and optimization.

Benefits of technology

It enables predictive maintenance of elevator malfunctions, improving elevator safety and reliability, reducing maintenance costs, and increasing the efficiency and quality of maintenance work.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118545590B_ABST
    Figure CN118545590B_ABST
Patent Text Reader

Abstract

This invention discloses a predictive maintenance method and device for elevators based on the Internet of Things (IoT) and machine learning, belonging to the field of public safety technology for special equipment. The method predicts potential elevator malfunctions through real-time monitoring and analysis of elevator operation data, allowing for proactive maintenance and upkeep. IoT technology enables real-time data acquisition and transmission from elevator equipment, providing a rich data source for machine learning. By collecting various parameters and status information during elevator operation, machine learning algorithms can deeply uncover hidden patterns and trends in the data, providing strong support for predictive maintenance. Through analysis and processing of this data, machine learning algorithms predict potential elevator malfunctions, issuing early warnings so maintenance personnel can intervene promptly to prevent malfunctions, thus improving elevator safety and reliability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention discloses a predictive maintenance method and device for elevators based on the Internet of Things and machine learning, which relates to the field of public safety technology for special equipment. Background Technology

[0002] With the acceleration of urbanization, elevators, as an important means of transportation in modern buildings, are seeing a continuous increase in their number and usage frequency. Consequently, related maintenance issues are gradually emerging, becoming a key factor affecting the safe and stable operation of elevators. Currently, the elevator maintenance industry faces numerous challenges, and predictive maintenance, as a new maintenance model, is gradually gaining attention and importance within the industry.

[0003] The current state of elevator maintenance is mainly manifested in the following aspects: First, the workload of maintenance is large and the tasks are heavy. With the increase in the number of elevators, the number of devices that maintenance personnel need to be responsible for has increased accordingly. The maintenance cycle is short and the tasks are heavy, making it difficult to achieve precision and comprehensiveness in maintenance work. Second, the technical level of maintenance personnel varies greatly. Due to the large differences in the skill level and experience of maintenance personnel, the quality of maintenance is difficult to guarantee. Some maintenance personnel only focus on surface repair work and ignore potential problems inside the elevator, which poses a hidden danger to the safe operation of the elevator. Third, the maintenance cost is high. The traditional maintenance model usually adopts the method of regular inspection and fault repair, which is inefficient and costly. Summary of the Invention

[0004] This invention addresses the problems of existing technologies by providing a predictive maintenance method and device for elevators based on the Internet of Things and machine learning. The technical solution adopted is as follows: Firstly, a predictive maintenance method for elevators based on the Internet of Things and machine learning includes: S1, An elevator status monitoring system is built based on the Internet of Things, and sensors are deployed in the elevator to acquire elevator operation data through the sensors; S2, Based on the elevator operation data, perform data processing through preprocessing; S3, Based on the elevator operation data after data processing, establish a subset of elevator fault prediction features through machine learning; S4. Based on the elevator fault prediction feature subset, an elevator fault prediction model is established using the random forest algorithm to predict the health status and remaining life of elevator internal components. S5. Based on the elevator fault prediction model, training and optimization are performed using historical data and normal operation data.

[0005] In some implementations, S3 specifically includes: S31, Based on the processed elevator operation data, feature data is obtained by extracting and selecting features using machine learning algorithms; S32, Based on the feature data, establish a feature subset for elevator fault prediction by calculating feature indicators.

[0006] In some implementations, in S3, the feature data includes: operational data features, sensor data features, time features, and maintenance history features; The operational data features include elevator usage frequency, load conditions, and speed variation data.

[0007] The sensor data features include temperature, current, voltage, and vibration data.

[0008] The time characteristics include usage time and operating time period data; The maintenance history features include the last maintenance time and historical fault records.

[0009] In some implementations, S4 specifically includes: S41, Based on the elevator fault prediction feature subset, the set is divided into a training set and a test set using the train_test_split function; S42, create a RandomForestClassifier object and specify the number of decision trees (n_estimators); S43, train the ladder fault prediction model using the training data, make predictions using the test set, and obtain an evaluation performance report using the accuracy score and classification report.

[0010] Secondly, embodiments of the present invention provide an elevator predictive maintenance device based on the Internet of Things and machine learning, comprising: The data acquisition module is used to build an elevator status monitoring system based on the Internet of Things, deploy sensors in the elevator, and acquire elevator operation data through the sensors; The preprocessing module is used to process the elevator operation data through preprocessing. The feature extraction module is used to establish a subset of elevator fault prediction features based on the processed elevator operation data through machine learning. The model building module is used to build an elevator fault prediction model based on the elevator fault prediction feature subset and the random forest algorithm to predict the health status and remaining life of elevator internal components. The model optimization module is used to train and optimize the elevator fault prediction model using historical data and normal operation data.

[0011] In some implementations, the feature extraction module specifically includes: The data processing unit is used to extract and select features based on the processed elevator operation data using machine learning algorithms to obtain feature data. The subset establishment unit is used to establish a subset of elevator fault prediction features by calculating feature indicators based on the feature data.

[0012] In some implementations, the feature extraction module includes the following feature data: runtime data features, sensor data features, time features, and maintenance history features. The operational data features include elevator usage frequency, load conditions, and speed variation data.

[0013] The sensor data features include temperature, current, voltage, and vibration data.

[0014] The time characteristics include usage time and operating time period data; The maintenance history features include the last maintenance time and historical fault records.

[0015] In some implementations, the model building module specifically includes: The subset partitioning unit is used to divide the elevator fault prediction feature subset into a training set and a test set using the train_test_split function; The object creation module is used to create RandomForestClassifier objects and specify the number of decision trees (n_estimators). The report generation unit is used to train the ladder fault prediction model using training data, make predictions using the test set, and obtain an evaluation performance report through accuracy score and classification report.

[0016] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein when the one or more computer instructions are executed by the processor, they implement the method described in the first aspect above.

[0017] Fourthly, embodiments of the present invention provide a computer storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, it implements the method described in the first aspect.

[0018] One or more embodiments of the present invention can bring at least the following beneficial effects: This invention utilizes predictive maintenance by real-time monitoring and analysis of elevator operation data to predict potential elevator malfunctions and enable proactive maintenance. IoT technology allows for real-time data acquisition and transmission from elevator equipment, providing a rich data source for machine learning. By collecting various parameters and status information during elevator operation, machine learning algorithms can delve into hidden patterns and trends within the data, providing strong support for predictive maintenance. Through analysis and processing of this data, machine learning algorithms predict potential elevator malfunctions, issuing early warnings so maintenance personnel can intervene promptly to prevent failures, thus improving elevator safety and reliability. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of an elevator predictive maintenance method based on the Internet of Things and machine learning provided in an embodiment of the present invention; Figure 2 This is a block diagram of an elevator predictive maintenance device based on the Internet of Things and machine learning, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the optimization of the random forest algorithm for predictive maintenance of elevators provided in the embodiments of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention. Example 1:

[0022] Figure 1 A flowchart illustrating a predictive maintenance method for elevators based on the Internet of Things and machine learning is shown, such as... Figure 1 As shown, the elevator predictive maintenance method based on the Internet of Things and machine learning provided in this embodiment includes: S1, An elevator status monitoring system is built based on the Internet of Things, and sensors are deployed in the elevator to acquire elevator operation data through the sensors; S2, Based on the elevator operation data, perform data processing through preprocessing; S3, Based on the elevator operation data after data processing, establish a subset of elevator fault prediction features through machine learning; S4. Based on the elevator fault prediction feature subset, an elevator fault prediction model is established using the random forest algorithm to predict the health status and remaining life of elevator internal components. S5. Based on the elevator fault prediction model, training and optimization are performed using historical data and normal operation data.

[0023] When the method of the present invention is working, firstly according to S1, an elevator status monitoring system is constructed based on Internet of Things technology. Various sensors are deployed in the key components and operation links of the elevator to collect various data in real time during the elevator operation process and capture key parameters during the elevator operation process, including but not limited to door opening and closing status, running speed, number of floor stops, load capacity, motor temperature, vibration, sound, current and other parameters. The system also collects elevator maintenance records, historical fault data and elevator operating environment information, and transmits these scattered data to the elevator status monitoring system in real time. Next, following S2, preprocess the collected data, including but not limited to cleaning, standardization, noise reduction, missing value imputation, normalization, and dataset partitioning, to ensure the accuracy and consistency of the data and provide reliable data support for subsequent model training.

[0024] The next step, according to S3, specifically includes: S31, Based on the processed elevator operation data, feature data is obtained by extracting and selecting features using machine learning algorithms; S32, Based on the feature data, establish a feature subset for elevator fault prediction by calculating feature indicators.

[0025] For the preprocessed data, machine learning algorithms are used for feature extraction and selection. By calculating indicators such as data statistics, correlation, information content, and elevator operation mechanism, a subset of features that have a significant impact on elevator fault prediction is selected. These features will serve as important inputs for subsequent model training.

[0026] Specifically, the feature data includes: operational data features, sensor data features, time features, and maintenance history features; The operational data features include elevator usage frequency, load conditions, and speed variation data.

[0027] Elevator usage frequency: Statistics on the number of times the elevator runs each day and the duration of each run can reflect the intensity of elevator usage.

[0028] Load conditions: including average load, maximum load, load distribution, etc., these data can reflect the load pressure of the elevator at different time periods.

[0029] Speed ​​changes: The speed changes during elevator start-up, acceleration, deceleration, and stopping can reflect the smoothness of elevator operation and whether there are any abnormalities.

[0030] The sensor data features include temperature, current, voltage, and vibration data.

[0031] Temperature: The temperature of critical components such as motors and control systems. Excessive temperature may indicate potential malfunctions.

[0032] Current and voltage: Current and voltage values ​​of elevator motors and other electrical components. Abnormal values ​​may indicate electrical faults.

[0033] Vibration: Vibration information during elevator operation is obtained through acceleration sensors. Excessive vibration may indicate a problem with mechanical components.

[0034] The time characteristics include usage time and operating time period data; Usage time: The total operating time of the elevator since its installation, which can reflect the degree of wear and tear on the elevator.

[0035] Operating time period: Analyzing elevator operating data at different time periods, such as peak hours and off-peak hours, helps to identify time-related failure modes.

[0036] The maintenance history features include the last maintenance time and historical fault records.

[0037] Last maintenance time: Records the time of the last elevator maintenance, which can reflect the elevator's maintenance status.

[0038] Historical fault records: The elevator's past fault records, including fault type, occurrence time, etc., are very valuable for predicting future faults.

[0039] The next step, according to S4, specifically includes: S41, Based on the elevator fault prediction feature subset, the set is divided into a training set and a test set using the train_test_split function; S42, create a RandomForestClassifier object and specify the number of decision trees (n_estimators); S43, train the ladder fault prediction model using the training data, make predictions using the test set, and obtain an evaluation performance report using the accuracy score and classification report.

[0040] A random forest algorithm is constructed to build an elevator failure prediction model to predict the health status and remaining lifespan of various elevator components. The dataset is divided into training and test sets using the train_test_split function. A RandomForestClassifier object is created, and the number of decision trees (n_estimators) is specified. The model is trained using the training data, and predictions are made on the test set. The model's performance is evaluated using the accuracy score and classification report.

[0041] The next step, following S5, is to train and optimize the model using historical fault data and normal operation data, such as... Figure 3 The diagram shown illustrates the optimization of the random forest algorithm for predictive elevator maintenance. This step enables the model to learn and identify different types of fault modes and predict future faults. As elevator operation data accumulates, the model is periodically updated and retrained to adapt to changes in elevator conditions and the emergence of new fault modes. Cross-validation, AUC, and accuracy metrics are used to evaluate and optimize model performance, ensuring the reliability and accuracy of the model in predicting elevator faults. Figure 1 As shown, the random forest algorithm is used to determine whether the elevator needs maintenance in the next 15 days based on the collected elevator data. Example 2:

[0042] Building upon Example 1, the trained prediction model established in step S1 is used to monitor and analyze the elevator's real-time operating data. When the model predicts a fault risk and triggers an early warning mechanism, it sends fault warning information, including fault type, fault location, and estimated fault time, to maintenance personnel in advance. The fault probability and type information output by the model assists maintenance personnel in fault diagnosis, improving the efficiency and accuracy of fault handling.

[0043] Based on the results of the predictive model and the actual operating conditions of the elevator, targeted maintenance plans are developed. For example, components predicted to be nearing the end of their lifespan are scheduled for replacement in advance; components that may experience accelerated wear are scheduled for enhanced inspection or preventative maintenance; and for elevators in good operating condition, the maintenance cycle can be appropriately extended to reduce maintenance costs. Based on the elevator's failure history and prediction results, maintenance resources are optimized to improve the efficiency and quality of maintenance work.

[0044] Maintenance strategies should be dynamic, adapting to changes in the elevator's actual operating conditions and forecast results. Predictive maintenance methods can dynamically adjust and optimize these strategies based on real-time monitoring data and forecasts, ensuring rational resource allocation, reducing operating costs, and improving the relevance and effectiveness of maintenance work. Data and experience gained during maintenance are fed back into the predictive model for updates and optimization, creating a virtuous cycle between maintenance work and model prediction.

[0045] After completing the maintenance task, the actual maintenance effect data is fed back to the predictive model to continuously update the model parameters, optimize the model's predictive capabilities, and form a data-driven predictive maintenance closed-loop management. Example 3:

[0046] Figure 2 A block diagram of a predictive elevator maintenance device based on the Internet of Things and machine learning is shown, such as... Figure 2 As shown, the elevator predictive maintenance device based on the Internet of Things and machine learning provided in this embodiment includes: The data acquisition module is used to build an elevator status monitoring system based on the Internet of Things, deploy sensors in the elevator, and acquire elevator operation data through the sensors; The preprocessing module is used to process the elevator operation data through preprocessing. The feature extraction module is used to establish a subset of elevator fault prediction features based on the processed elevator operation data through machine learning. The model building module is used to build an elevator fault prediction model based on the elevator fault prediction feature subset and the random forest algorithm to predict the health status and remaining life of elevator internal components. The model optimization module is used to train and optimize the elevator fault prediction model using historical data and normal operation data.

[0047] Furthermore, the feature extraction module specifically includes: The data processing unit is used to extract and select features based on the processed elevator operation data using machine learning algorithms to obtain feature data. The subset establishment unit is used to establish a subset of elevator fault prediction features by calculating feature indicators based on the feature data.

[0048] Furthermore, in the feature extraction module, the feature data includes: operational data features, sensor data features, time features, and maintenance history features; The operational data features include elevator usage frequency, load conditions, and speed variation data.

[0049] The sensor data features include temperature, current, voltage, and vibration data.

[0050] The time characteristics include usage time and operating time period data; The maintenance history features include the last maintenance time and historical fault records.

[0051] Furthermore, the model building module specifically includes: The subset partitioning unit is used to divide the elevator fault prediction feature subset into a training set and a test set using the train_test_split function; The object creation module is used to create RandomForestClassifier objects and specify the number of decision trees (n_estimators). The report generation unit is used to train the ladder fault prediction model using training data, make predictions using the test set, and obtain an evaluation performance report through accuracy score and classification report. Example 4:

[0052] This embodiment also provides an electronic device, including a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method of Embodiment 1; In practical applications, the processor can be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller unit (MCU), microprocessor, or other electronic components to execute the methods described in the above embodiments.

[0053] The method implemented in this embodiment is as shown in Embodiment 1. Example 5:

[0054] This embodiment also provides a computer storage medium, in which a computer program is stored, and when the computer program is executed by one or more processors, it implements the method of embodiment one. The computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0055] The method implemented in this embodiment is as shown in Embodiment 1.

[0056] In the several embodiments provided in this invention, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system and method embodiments described above are merely illustrative.

[0057] It should be noted that, in this document, the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0058] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A predictive maintenance method for elevators based on the Internet of Things and machine learning, characterized in that, include: S1, An elevator status monitoring system is built based on the Internet of Things, and sensors are deployed in the elevator to acquire elevator operation data through the sensors; S2, Based on the elevator operation data, perform data processing through preprocessing; S3, Based on the elevator operation data after data processing, establish a subset of elevator fault prediction features through machine learning; S4. Based on the elevator fault prediction feature subset, an elevator fault prediction model is established using the random forest algorithm to predict the health status and remaining life of elevator internal components. S5. Based on the elevator fault prediction model, training and optimization are performed using historical data and normal operation data. S4 specifically includes: S41, Based on the elevator fault prediction feature subset, the set is divided into a training set and a test set using the train_test_split function; S42, create a RandomForestClassifier object and specify the number of decision trees (n_estimators); S43, the elevator fault prediction model is trained using training data, predictions are made using the test set, and an evaluation performance report is obtained through accuracy and classification reports.

2. The method according to claim 1, characterized in that, S3 specifically includes: S31, Based on the processed elevator operation data, feature data is obtained by extracting and selecting features using machine learning algorithms; S32, Based on the feature data, establish a feature subset for elevator fault prediction by calculating feature indicators.

3. The method according to claim 2, characterized in that, In S3, the feature data includes: operational data features, sensor data features, time features, and maintenance history features; The operational data features include elevator usage frequency, load conditions, and speed variation data; The sensor data features include temperature, current, voltage, and vibration data; The time characteristics include usage time and operating time period data; The maintenance history features include the last maintenance time and historical fault records.

4. An elevator predictive maintenance device based on the Internet of Things and machine learning, characterized in that, include: The data acquisition module is used to build an elevator status monitoring system based on the Internet of Things, deploy sensors in the elevator, and acquire elevator operation data through the sensors; The preprocessing module is used to process the elevator operation data through preprocessing. The feature extraction module is used to establish a subset of elevator fault prediction features based on the processed elevator operation data through machine learning. The model building module is used to build an elevator fault prediction model based on the elevator fault prediction feature subset and the random forest algorithm to predict the health status and remaining life of elevator internal components. The model optimization module is used to train and optimize the elevator fault prediction model using historical data and normal operation data.

5. The apparatus according to claim 4, characterized in that, The feature extraction module specifically includes: The data processing unit is used to extract and select features based on the processed elevator operation data using machine learning algorithms to obtain feature data. The subset establishment unit is used to establish a subset of elevator fault prediction features by calculating feature indicators based on the feature data.

6. The apparatus according to claim 5, characterized in that, In the feature extraction module, the feature data includes: operational data features, sensor data features, time features, and maintenance history features; The operational data features include elevator usage frequency, load conditions, and speed variation data; the sensor data features include temperature, current, voltage, and vibration data. The time characteristics include usage time and operating time period data; The maintenance history features include the last maintenance time and historical fault records.

7. The apparatus according to claim 6, characterized in that, The model building module specifically includes: The subset partitioning unit is used to divide the elevator fault prediction feature subset into a training set and a test set using the train_test_split function; The object creation module is used to create RandomForestClassifier objects and specify the number of decision trees (n_estimators). The report generation unit is used to train the elevator fault prediction model using training data, make predictions using the test set, and obtain an evaluation performance report based on accuracy and classification reports.

8. An electronic device, characterized in that, The system includes a memory and a processor, the memory being used to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, implement the method as described in any one of claims 1-4.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a computer program that, when executed by a processor, is used to implement the method as described in any one of claims 1-4.

Citation Information

Patent Citations

  • Elevator fault pre-diagnosis method

    CN116812698A